linux-kernel - Re: [PATCH 5/9] drm/meson: dw-hdmi: split resets out of hw init.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAFBinCCvWFCCvb9gPvv0-eudG=iuKROk5rPSiorKTnHcToDfTQ@mail.gmail.com>
Date: Tue, 6 Aug 2024 22:49:40 +0200
From: Martin Blumenstingl <martin.blumenstingl@...glemail.com>
To: Jerome Brunet <jbrunet@...libre.com>
Cc: Neil Armstrong <neil.armstrong@...aro.org>, 
	Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>, Maxime Ripard <mripard@...nel.org>, 
	Thomas Zimmermann <tzimmermann@...e.de>, David Airlie <airlied@...il.com>, Daniel Vetter <daniel@...ll.ch>, 
	Kevin Hilman <khilman@...libre.com>, dri-devel@...ts.freedesktop.org, 
	linux-amlogic@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/9] drm/meson: dw-hdmi: split resets out of hw init.

Hi Jerome,

On Tue, Jul 30, 2024 at 2:50 PM Jerome Brunet <jbrunet@...libre.com> wrote:
>
> This prepares the migration to regmap usage.
>
> To properly setup regmap, the APB needs to be in working order.
> This is easier handled if the resets are not mixed with hw init.
>
> More checks are required to determine if the resets are needed
> on resume or not. Add a note for this.
>
> Signed-off-by: Jerome Brunet <jbrunet@...libre.com>
> ---
>  drivers/gpu/drm/meson/meson_dw_hdmi.c | 14 +++++++++-----
>  1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c
> index 5cd3264ab874..47aa3e184e98 100644
> --- a/drivers/gpu/drm/meson/meson_dw_hdmi.c
> +++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c
> @@ -581,11 +581,6 @@ static void meson_dw_hdmi_init(struct meson_dw_hdmi *meson_dw_hdmi)
>         /* Bring HDMITX MEM output of power down */
>         regmap_update_bits(priv->hhi, HHI_MEM_PD_REG0, 0xff << 8, 0);
>
> -       /* Reset HDMITX APB & TX & PHY */
> -       reset_control_reset(meson_dw_hdmi->hdmitx_apb);
> -       reset_control_reset(meson_dw_hdmi->hdmitx_ctrl);
> -       reset_control_reset(meson_dw_hdmi->hdmitx_phy);
> -
>         /* Enable APB3 fail on error */
>         if (!meson_vpu_is_compatible(priv, VPU_COMPATIBLE_G12A)) {
>                 writel_bits_relaxed(BIT(15), BIT(15),
> @@ -675,6 +670,10 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master,
>                 return PTR_ERR(meson_dw_hdmi->hdmitx_phy);
>         }
>
> +       reset_control_reset(meson_dw_hdmi->hdmitx_apb);
> +       reset_control_reset(meson_dw_hdmi->hdmitx_ctrl);
> +       reset_control_reset(meson_dw_hdmi->hdmitx_phy);
The old out of tree vendor driver [0] enables the "isfr" and "iahb"
(in P_HHI_HDMI_CLK_CNTL and P_HHI_GCLK_MPEG2) clocks before triggering
the resets.
Previously meson_dw_hdmi's behavior was identical as it enabled the
clocks in meson_dw_hdmi_bind() and only later triggered the resets.

I'm totally fine with moving the resets to meson_dw_hdmi_bind() but I
think it should happen after devm_clk_bulk_get_all_enable() has been
called (to keep the order and thus avoid side-effects that we don't
know about yet).

Also out of curiosity: are you planning to convert the driver to use
devm_reset_control_bulk_get_exclusive()?


Best regards,
Martin


[0] https://github.com/endlessm/linux-s905x/blob/master/drivers/amlogic/hdmi/hdmi_tx_20/hw/hdmi_tx_hw.c#L470